Cross-Vendor Multi-Node Spot Instance Dynamics: A Public Cloud Data Archive Service and Analysis

This repository contains a comprehensive analysis of cloud spot instance data collected from three major cloud service providers: AWS, Azure, and GCP. The dataset includes over 160 million records collected during June 2025, covering spot pricing, interruption frequencies, placement scores, and various stability metrics across 7 major regions per provider.The analysis examines temporal patterns (hourly and weekly variations), geographical differences (regional and availability zone-level), instance category characteristics (general, compute, memory, storage, and accelerated computing), and statistical relationships between key metrics including spot price, interruption frequency (IF), spot placement score (SPS), and advanced availability metrics (T2, T3).This dataset and analysis pipeline are designed to support cloud cost optimization research, reliability engineering studies, capacity planning, and broader research into cloud resource pricing and allocation mechanisms. All analysis scripts are provided in both Python and Jupyter notebook formats, with generated visualizations and statistical tables included in the repository.